JP2006126919A - Data management device for flash memory and control method of flash memory - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable restart, in the event of power supply interruption during garbage collection, from a point where abnormality occurred at the next starting. <P>SOLUTION: A flash memory 9 is composed of a data unit, a data management unit and a spare unit. A block copy flag is provided on the unit header of each unit, and the data management unit includes entry information, a sector copy flag, an entry copy flag and a validity/invalidity determination part. In recovery processing, whether garbage collection is in execution or not and whether copying is performed for the data unit or not are determined based on the value of the block copy flag. The previous interruption position is discriminated from the state of the sector copy flag, and copying is continued from this position. Since the previous interruption position can be found out from the state of the entry copy flag, copying is continued from this position. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an electronic device including a flash memory, and more particularly to a data management device for a flash memory and a control method for the flash memory.

Conventionally, access control including data writing in block units and batch data erasing in sector units is performed based on sector management information and block management information, and a sector to which data is to be written is selected based on the number of sector erasures A flash memory control method is known (for example, see Patent Document 1). The flag values “sector erase start (00000)”, “sector erase in progress (11111)”, and “garbage collection in progress (00111)” are in the middle of processing the sector data, and the authenticity of the data is not clear. The contents of the block data portion of the sector are erased again.
JP 2000-330850 A

  However, in the control method disclosed in Patent Document 1, when garbage collection is activated, the state is temporarily changed to “sector erase start”, “sector erase in progress”, and “garbage collection in progress” states. If a power loss (power interruption) occurs in this state, the authenticity of the data is not clear, so even a sector for which data copying has been completed must be erased again. This results in a sector erasure that takes a non-short time for the system. In addition, the number of sector erasures in the flash memory is increased, and there is a problem that the lifetime of the device is shortened.

  The present invention has been made in consideration of such circumstances, and the purpose of the present invention is to provide a flash memory that can continue from the location where an abnormality occurs at the next startup even if a power failure occurs during garbage collection. A data management apparatus and a flash memory control method are provided.

  The present invention is a data management device for a flash memory comprising a plurality of data units for accessing data for each sector unit and a data management unit for storing entry information of data stored in the data units. , An access status recording means for recording a data access status during a garbage collection operation, and data for recognizing a data access interruption position based on the data access status at the time of system startup and continuing data access from the interruption position And an access control means.

  According to the present invention, the access status recording means records a sector copy interruption position during a garbage collection operation as a value of a sector copy flag provided in the data management unit, and the data access control means At the time of the above, a sector copy interruption position is recognized based on the value of the sector copy flag, and sector copying is continued from the interruption position.

  According to the present invention, the access status recording means records an entry copy interruption position during a garbage collection operation as a value of an entry copy flag provided in the data management unit, and the data access control means At the time of the above, based on the value of the entry copy flag, the entry copy interruption position is recognized, and the entry copy is continued from the interruption position.

  The present invention further comprises block copy control means for recording a copy state between data units during garbage collection as a value of a block copy flag provided in the data unit header, wherein the data access control means comprises the block Based on the value of the copy flag, it is determined whether to continue sector copying from the interruption position based on the value of the sector copy flag or whether to continue entry copying from the interruption position based on the value of the entry copy flag. It is characterized by that.

  The present invention also provides a flash memory control method comprising a plurality of data units that access data for each sector unit, and a data management unit that stores data entry information stored in the data units. The data access situation during the garbage collection operation is recorded, and when the system is started, the data access interruption position is recognized based on the data access situation, and the data access is continued from the interruption position.

  Further, the present invention records the sector copy interruption position during the garbage collection operation as the value of the sector copy flag provided in the data management unit. Based on the value of the sector copy flag at the time of system startup, It is characterized by recognizing a copy interruption position and continuing sector copying from the interruption position.

  Further, the present invention records the entry copy interruption position during the garbage collection operation as the value of the entry copy flag provided in the data management unit. Based on the value of the entry copy flag when the system is started up, It is characterized by recognizing the copy interruption position and continuing the entry copy from the interruption position.

  Further, the present invention records a copy state between data units during garbage collection as a value of a block copy flag provided in the data unit header, and based on the value of the block copy flag, the value of the sector copy flag Whether to continue sector copying from the interruption position or to determine whether to continue entry copying from the interruption position based on the value of the entry copy flag.

  According to the present invention, the access status recording means records the data access status during the garbage collection operation, and the data access control means recognizes the data access interruption position based on the data access status when the system is started up. Since the data access is continued from the interruption position, even if the power is cut off during the garbage collection, it is possible to continue from the location where the abnormality occurs at the next activation.

Hereinafter, an electronic apparatus according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of the embodiment of the present invention. Here, a mobile phone terminal will be described as an example of an electronic device. In FIG. 1, reference numeral 1 denotes a wireless unit that establishes a communication line with a base station and performs communication. Reference numeral 2 denotes a control unit that performs overall control of the operation of the mobile phone terminal. Reference numeral 3 denotes a RAM that temporarily stores user-specific information and data, and a ROM that stores programs for operating the mobile phone terminal in advance. Reference numeral 5 denotes a microphone that collects the call voice and a speaker that generates the call voice. Reference numeral 6 denotes a key operation unit including dial keys, function keys, and the like. Reference numeral 7 denotes a display unit composed of a liquid crystal display. Reference numeral 8 denotes a flash memory driver for accessing data to the flash memory 9.

  Here, the configuration of the storage area in the flash memory 9 shown in FIG. 1 will be described with reference to FIG. FIG. 2 is an explanatory diagram showing the configuration of the storage area of the flash memory 9 shown in FIG. The flash memory 9 is divided into units for each block erase unit. Here, it is assumed that each unit is composed of six data units, one data management unit and one spare unit, and each unit has a capacity of 8 kbytes. The data unit is a unit for storing data as the name indicates, and each data unit has sectors separated by 64 bytes. This sector is a writing unit as data.

  The data management unit is a unit that manages data units and has entry information in units of 8 bytes. In addition, a unit header is provided for the first one sector of each unit. This unit header is provided with a unit identification mark, a block erase counter, and a block copy flag which will be described later. The unit identification mark is used to identify whether the unit is a data management unit, a spare unit, or a data unit. This mark is rewritten each time a predetermined block is used depending on the use at that time. At the time of formatting, the data management unit is assigned to the first block, and the spare unit is assigned to the last block. The block erase counter records the number of times the block has been erased, and is set to 00h during formatting. Each unit header stores a value from 00h to FFh, and returns to 00h after FFh by control.

  The spare unit is a unit used when executing garbage collection or wear leveling. The entry information of the data management unit includes a 64-byte sector start address, unit number (block number for data storage), sector number, sector copy flag, entry copy flag, valid / invalid discriminator, dirty / clean of invalid entry It has a discrimination part. The valid / invalid determination unit overwrites 00h when data is deleted. Although the data is invalid, at this point, the dirty / clean discriminator of the invalid entry is still dirty. That is, it indicates an invalid sector managed on the flash memory system. The dirty / clean discriminating unit overwrites 00h when garbage is collected. That is, the data is invalid and clean. For data management, it is completely deleted. The entry is updated by writing the new entry into the empty area and invalidating the old entry. When the data management unit including the entry information is garbage collected, the dirty entry is copied to the spare unit, but the clean entry unit is already garbage collected, so the clean entry is not copied.

  Next, garbage collection will be described. Garbage collection means that erased / edited data causes an invalid sector in the data unit. Except for this, only the valid sector is copied to the spare unit, and the source unit is block erased. The unit is to be the next spare unit. At the same time as copying, the entry information of the data management unit is also updated by writing new entry information and invalidating the old entry information. The target unit for garbage collection is the unit with the most invalid sectors. Also, the garbage collection start condition is when it is determined that no unit is empty at the time of writing (before writing), or when the number of invalid sectors exceeds a predetermined threshold at the time of completion of writing / waiting.

The procedure for executing garbage collection is as follows.
(1) Search the data management unit and select a target block for garbage collection.
(2) Copy the valid sector to the spare unit.
(3) Write new entries sequentially in conjunction with (2) above.
(4) The old entries are invalidated sequentially in conjunction with (2) and (3) above.
(5) Erase unnecessary blocks.

  The data management unit is also subject to garbage collection, and the activation condition is when it is determined that the number of empty entries that can be added to the data management unit is less than a certain value. In the operation, valid / invalid sectors on the entry are copied to the spare unit, and the clean part is not copied. Then, the block of the original entry that is no longer needed is erased and used as the next spare unit.

The procedure for executing data collection garbage collection is as follows.
(1) It is determined that the number of empty entries that can be added to the data management unit is less than a certain value.
(2) The valid / invalid sector on the entry is copied to the spare unit (the clean part is not copied).
(3) The entry copy flag is sequentially operated in conjunction with the above (2).
(4) Erase the original block that is no longer needed.

  Next, wear leveling will be described. Wear leveling means that in the flash memory 9 in which the number of erasures is limited, all the blocks are managed so as to be used on average, so that only a specific block is erased repeatedly and the lifetime ends first. In order to avoid this, the entire memory is effectively used. For activation of wear leveling, each block erase counter is checked, and if it is equal to or greater than the threshold for wear leveling activation, the block having the smallest counter value is selected. Then, all the data in this block is copied to the spare unit, at the same time, new entry information is written, the old entry information is invalidated, and finally the copy source is erased to make the next spare unit. Wear leveling is not executed until the difference between each block erase counter value (minimum and maximum) exceeds the threshold for starting wear leveling.

The procedure for wear leveling is as follows.
(1) Check each block erase counter. If the block erase counter is equal to or greater than the wear leveling activation threshold, select the block with the smallest counter value.
(2) Copy all data in the block to the spare unit.
(3) Since the copy source and copy destination unit numbers are known, the new entry is written in conjunction with the above (2), and the old entry is invalidated.
(4) Block erasing the copy source.

  Priorities are set for garbage collection or wear leveling that is activated at the end of writing / standby. The first is garbage collection of the data management unit, the second is garbage collection of the data unit, and the third is wear leveling. In one activation, any one is executed according to the priority order.

  Next, the configuration of the flash memory driver 8 shown in FIG. 1 will be described with reference to FIG. FIG. 3 is a block diagram showing a configuration of the flash memory driver 8 shown in FIG. The application I / F unit 81 receives a write / read request from the application side and sends it to the sequence control unit 84. The sequence control unit 84 executes and controls the access unit by sequentially performing an operation desired to be performed on the flash memory, that is, a request operation from the application side. The unit header access unit 85 is for accessing the unit header, and writes / reads a unit identification mark, a block erase counter, and a block copy flag. Actual access is performed by executing a device I / F unit 89 described later.

  The entry access unit 86 writes / reads a unit number, sector number, valid / invalid discrimination unit, and the like. Actual access is performed by executing a device I / F unit 89 described later. The sector data access unit 87 is for writing / reading data of a 64-byte sector. The entry search unit 88 searches for entry information such as searching for a data management unit to obtain a unit having the largest number of invalid sectors. The device I / F unit 89 directly executes writing / reading with respect to the flash memory.

  Next, a mechanism that enables the flash memory system of this embodiment to be recovered will be described. When the terminal system starts up, it is checked whether the contents of the flash memory are normal. If it is not normal, recovery processing is executed to perform recovery / restoration.

  If the power interruption occurs during garbage collection during the previous use, it is confirmed that the contents of the flash memory are invalid when the terminal power is next activated, and garbage collection is resumed and completed. Then, the data management unit and the data part are returned to the normal state.

  The block copy flag provided in the unit header is used to control block copy at the time of garbage collection or wear leveling. This block copy flag is normally FFh, but is used for overwriting when a block copy occurs. The value of the block copy flag and the state at that time are as follows.

FFh: A unit not involved in block copy.
1Eh: A block copy destination unit corresponding to a spare unit.
16h: When copying is completed, 1Eh of the copy destination unit is overwritten with 16h.
06h: After erasing (erasing) the copy source unit, the copy destination unit is changed from 16h to 06h.
04h: Block copy source unit (in the case of a data management unit).
02h: Block copy source unit (in the case of a data unit).

The order of setting the block copy flag is
(1) The spare unit is changed from FFh to 1Eh.
(2) When the copy source unit is a data unit, it is set to 02h, and when it is a data management unit, it is set to 04h (copying is started after overwriting on 02h / 04h).
(3) The spare unit is changed from 1Eh to 16h (overwriting to 16h when copying of all data of the unit is completed).
(4) Erase (erase) the copy source unit.
(5) The copy destination unit is changed from 16h to 06h.

  FIG. 4 is an explanatory diagram showing an example of control of the block copy flag. In the figure, the copy source is described as “FF / 06”, but “FF” indicates that the flash memory is in the initial state after erasure, and “06” is once stored in the copy destination block in the previous garbage collection. It shows that it has become.

  Recovery when a power failure occurs while the block copy flag of the copy destination spare unit is set to 1 Eh does not need to be performed because the copy has not yet started. After the block copy flag is set to 1Eh, the same value is overwritten when it is used next time.

  Recovery when a power interruption occurs while copying is in progress can check how far the data has been copied by checking the entry information of the data management unit. Thereby, the subsequent processing can be executed. The sector copy flag is checked when the data unit is being copied, and the entry copy flag is checked when the data management unit is being copied.

  Recovery when a power interruption occurs in the copy completed state recognizes that the block that has been copied but has been invalidated has not yet been erased. Therefore, the copy source block, that is, the block whose block copy flag is 02h or 04h can be erased.

  In the recovery when the power interruption occurs in the state after erasing the copy source block, first, since the block copy flag of the copy destination is 16h, the block whose copy source is 02h or 04h is searched first. However, since the copy source is after erasure, there is no block copy flag of 02h or 04h, and it is possible to cope with recovery only by recognizing that the block has already been erased and setting the copy destination block copy flag to 06h.

  The sector copy flag is controlled at the time of copying the data unit. The sector copy flag is present in each entry in the data management unit. The value of the sector copy flag and the state at that time are as follows.

FFh: State before writing from one sector of the copy source to the flash memory.
0Fh: Set to 0Fh immediately before writing to the flash memory. It can be considered that the flash memory is being written.
03h: Set to 03h when all bytes have been written (copied) to the flash memory.
00h: After invalidating the old entry, the sector copy flag of this new entry is set to 00h.

  Here, FIG. 5 is an explanatory diagram showing an example of control of the sector copy flag. In the figure, the old entry is described as “00 / FF”, but “FF” indicates that the flash memory is in the initial state after erasure, and “00” once becomes a new entry in the previous garbage collection. It shows that.

  When a power failure occurs while writing, the new entry whose sector copy flag is 0Fh (copying to the sector indicated by the entry) is invalidated, a new entry is newly created, and copying is resumed.

  For recovery when the power is cut off in the write end state, two entries are valid at this time. However, the sector copy flag is different and is 00h for the old entry and 03h for the new entry. That is, it is detected that the sector copy flag is 03h, the address is known, an old entry with the same address can be recognized, and it can be invalidated.

  Recovery when a power failure occurs in the state after invalidating the old entry only needs to set the sector copy flag of the new entry to 00h.

  When the data management unit is being copied, the entry copy flag is controlled. The entry copy flag is a flag existing in each entry in the data management unit, and is operated on the copy source entry or the copy destination entry. The value of the entry copy flag and the state at that time are as follows.

FFh: State before copying.
0Fh: The old entry is set to 0Fh immediately before the entry is written to the copy destination. Write starts.
CFh: The new entry is set to CFh immediately before the entry is written to the copy destination. It can be considered that the flash is being written.
8Fh: The new entry is set to 8Fh after the entry write to the copy destination is completed.

  Here, FIG. 6 is an explanatory diagram showing an example of control of the entry copy flag. In the figure, the old entry is described as “FF / 8F”, but “FF” indicates that the flash memory is in the initial state after erasure, and “8F” is a new entry once in garbage collection. Is shown.

  In the recovery when the power interruption occurs in the state before copying, the old entry last set to 0Fh is searched, and if there is, the next entry is copied.

  The recovery when the power interruption occurs in the write start state is immediately after the old entry is set to 0Fh, and since it has not been written yet, the new entry is written to the empty termination area of the copy destination.

  Recovery when a power failure occurs while the data is being written invalidates the halfway new entry (in this case, it is strictly necessary to clean the new entry).

  The recovery when the power is cut off in the write end state has completed the write, and the process proceeds to the next entry copy.

  As a supplement, write (copy) is performed after the entry copy flag of the old entry is set to 0Fh. However, if the power is turned off immediately before writing (that is, the flag value on the new entry side is FFh), “true” writing is in progress. Absent. That is, the unit number of the new entry is FFh. At this time, there is no new entry, so nothing should be invalidated. If the power is cut off during writing (that is, the flag value on the new entry side is CFh), the unit number is invalidated because it is written halfway rather than FFh.

  Regarding recovery when the power is cut off during garbage collection, it is as follows. If the power is cut off during data copy by garbage collection, the new entry whose sector copy flag is 0Fh (copying to the sector indicated by the entry) is invalidated at the next power-on, a new entry is recreated and copying is resumed. First, the block copy flag in the unit header is checked to confirm the copy source block (block copy flag: 02h) and the copy destination block (block copy flag: 1Eh). During garbage collection, new and old entries are updated in conjunction with sector copying, so it is possible to determine the sector to be copied continuously by checking how far the entry of the copy source unit has been invalidated.

  If the power is cut off during the garbage collection of the data management unit, the next entry is copied between units at the next power-on to complete the garbage collection. First, the block copy flag in the unit header is checked to confirm the copy source block (block copy flag: 04h) and the copy destination block (block copy flag: 1Eh). The entry copy flag indicates how far the entry of the copy source unit has been copied (check 0Fh). Copying for one entry (8 bytes) is executed halfway, and for entries that have not been copied, the incomplete entry in the copy destination unit is invalidated (cleaned) and then copied to the next area. When copying is completed up to the final entry, the copy source unit is erased to complete garbage collection.

  When the power is turned off during wear leveling, the processing is the same as when the power is turned off during garbage collection.

  Next, a specific example of the flash memory system recovery method will be described. The flash memory system recovery process is also executed during the initialization process executed when the terminal power supply is activated. Here, FIG. 7 to FIG. 9 are flowcharts for explaining the recovery processing according to the present embodiment.

  First, the block copy flag (BCF) is read (S10), and it is determined whether or not there is a unit whose block copy flag in the unit header is 02h or 04h (S12). Here, in the case where such a unit exists, it means that a power interruption has occurred during garbage collection. If no such unit exists, the process returns to the normal process.

  On the other hand, if there is a unit whose block copy flag is 02h or 04h, it is determined whether the data unit is being copied, that is, whether the block copy flag is 02h or 04h (S14). If the block copy flag is 02h, the data unit garbage collection restart recovery process is executed (S16). If the block copy flag is 04h, the data management unit garbage collection restart recovery process is executed. (S18).

  If the block copy flag is 02h in step S14 described above, the unit is copied from the 02h unit to the 1Eh unit as the recovery process. The extent to which sector copying has been completed can be determined by the sector copy flag status to what extent the entry having the copy source unit number of the data management unit has been invalidated. That is, since the sector copy interruption position is known, it is only necessary to copy the continuation from that position, and it is not necessary to erase the copy destination block again and copy from the beginning. Hereinafter, this will be described in detail with reference to FIG.

  In the garbage collection resumption recovery processing of the data unit shown in FIG. 8, first, it is determined whether or not the sector copy flag (SCF) is being written, that is, whether or not there is a sector copy flag in the data management unit of 0Fh. Judgment is made (S30), and if the data is being written, of the 64 bytes of data pointed to by the written entry, not all of the copying has been completed, so once the data in the data management unit corresponding to the interrupted sector is stored. The entry is invalidated (S32), a new entry is created (S34), and sector copying is repeated (S40).

  On the other hand, when there is no sector copy flag of 0Fh, it means the end of writing, so it is determined whether or not the old entry has been invalidated (S36). Is repeated (S40). If the old entry has not been invalidated, it is invalidated (S38), and sector copying is repeated (S40).

  If the sector copy flag is 0Fh and the unit number or sector number is FFh, the data entity does not exist as data on the flash. As a recovery process, it is only necessary to invalidate this entry of the data management unit. Existing entries are still valid.

  If the block copy flag is 04h in step S14 described above, it indicates that copying is being performed as garbage collection of the data management unit, and entry copying is resumed from the unit of 04h to the unit of 1Eh. The extent to which entry copying has been completed can be determined by looking at the entry copy flag. That is, since the entry copy interruption position is known, it is only necessary to copy the continuation from that position, and it is not necessary to erase the copy destination block again and copy from the beginning. Hereinafter, this will be described in detail with reference to FIG.

  In the garbage collection restart recovery process of the data management unit shown in FIG. 9, it is first determined whether or not the entry copy flag (ECF) is at the start of writing (S50). When the entry copy flag indicates the write start time, the new entry is written (S52), and the entry copy is repeated (S60). On the other hand, when the entry copy flag does not indicate the start of writing, it is determined whether or not the entry copy flag (ECF) indicates that writing is in progress (S54). When the entry copy flag indicates that writing is in progress, the suspended entry is invalidated (S56), a subsequent entry is created (S58), and entry copying is repeated (S60). If the entry copy flag does not indicate that writing is in progress, that is, indicates that writing has been completed, it means that one entry copy has been completed, so the remaining entry copies are repeated as they are (S60).

  Here, if the block copy flag is not 1Eh but 16h, the copying to the unit is completed, and the unit of 02h or 04h is erased as a block.

  According to the above-described embodiment, even if the flash memory is destroyed due to memory management due to an improper termination in the previous operation, the flash memory is normalized by executing recovery processing at the next terminal startup. can do. Even if the power is cut off during garbage collection, the block copy can be continued from the middle of occurrence of an abnormality, not from the beginning, and the startup time of the terminal can be increased. In addition, since it is only necessary to continue copying from the middle without executing garbage collection from the beginning, it is not necessary to erase the block written halfway, and it is not necessary to wait for the block erasing time. A negative impact on device life can be avoided.

  Further, the program for realizing the function of the processing unit in FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system and executed to control the flash memory 9. Processing may be performed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in the computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) in a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.

The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.
In addition, when the system is started up, the case where the power supply is activated has been described in the above-described embodiment, but the case where the system is reset by a reset operation is also included.

It is a block diagram which shows the structure of embodiment of this invention. 3 is an explanatory diagram showing a configuration of a storage area of a flash memory 9. FIG. 3 is a block diagram showing a configuration of a flash memory driver 8. FIG. It is explanatory drawing which shows an example of control of a block copy flag. It is explanatory drawing which shows an example of control of a sector copy flag. It is explanatory drawing which shows an example of control of an entry copy flag. It is a flowchart for demonstrating the recovery process by this embodiment. It is a flowchart for demonstrating the recovery process by this embodiment. It is a flowchart for demonstrating the recovery process by this embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Wireless part, 2 ... Control part, 3 ... Memory part, 5 ... Microphone / speaker, 6 ... Key operation part, 7 ... Display part, 8 ... Flash memory Driver 9, flash memory 81, application I / F section 84, sequence control section (data access control means) 85, unit header access section (access status recording means), 86 ..Entry access section (access status recording means), 87... Sector data access section (access status recording means), 88... Entry search section, 89... Device I / F section

Claims (8)

A data management device for flash memory comprising a data unit for accessing data for each sector unit and a data management unit for storing entry information of data stored in the data unit,
Access status recording means for recording the data access status during the garbage collection operation;
A data management device for flash memory, comprising: a data access control means for recognizing a data access interruption position based on the data access status at the time of system startup and continuing data access from the interruption position . The access status recording means records the sector copy interruption position during the garbage collection operation as a value of a sector copy flag provided in the data management unit,
2. The data access control means according to claim 1, wherein when the system is started up, the sector copy flag is recognized based on the value of the sector copy flag, and the sector copy is continued from the interrupt position. Data management device for flash memory. The access status recording means records the entry copy interruption position during the garbage collection operation as a value of an entry copy flag provided in the data management unit,
2. The data access control unit according to claim 1, wherein when the system is started up, the entry copy interruption position is recognized based on the value of the entry copy flag, and the entry copy is continued from the interruption position. Data management device for flash memory. Block copy control means for recording a copy status between data units during garbage collection as a value of a block copy flag provided in the data unit header;
The data access control means continues the sector copy from the interruption position based on the value of the sector copy flag based on the value of the block copy flag or starts from the interruption position based on the value of the entry copy flag. 2. The data management apparatus for flash memory according to claim 1, wherein it is determined whether or not to continue entry copying. A flash memory control method comprising a plurality of data units that access data for each sector unit and a data management unit that stores entry information of data stored in the data units,
Record the data access status during garbage collection,
A flash memory control method characterized by recognizing a data access interruption position based on the data access status at the time of system startup and continuing data access from the interruption position. The sector copy interruption position during the garbage collection operation is recorded as the value of the sector copy flag provided in the data management unit,
6. The flash memory control method according to claim 5, wherein when the system is started up, the sector copy interruption position is recognized based on the value of the sector copy flag, and sector copying is continued from the interruption position. The entry copy interruption position during the garbage collection operation is recorded as the value of the entry copy flag provided in the data management unit,
6. The flash memory control method according to claim 5, wherein when the power is turned on, the entry copy interruption position is recognized based on the value of the entry copy flag, and the entry copy is continued from the interruption position. The copy status between data units during garbage collection is recorded as the value of the block copy flag provided in the data unit header,
Based on the value of the block copy flag, it is determined whether to continue sector copy from the interrupt position based on the value of the sector copy flag or to continue entry copy from the interrupt position based on the value of the entry copy flag 6. The flash memory control method according to claim 5, wherein:

Images